Time difference of arrival-based three-station lightning locating system in Malaysia
Lightning is an electrical discharge that happens during thunderstorms. The discharge type can be either within cloud (IC), cloud-to-cloud, or cloud-to-ground (CG). The characterisation of a discharge activity based on the analysis of the measured electric field is widely studied in various geograph...
Saved in:
| Main Author: | |
|---|---|
| Format: | Thesis |
| Language: | English |
| Published: |
2015
|
| Subjects: | |
| Online Access: | http://eprints.utm.my/id/eprint/54808/1/BehnamSalimiPFKE2015.pdf |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Lightning is an electrical discharge that happens during thunderstorms. The discharge type can be either within cloud (IC), cloud-to-cloud, or cloud-to-ground (CG). The characterisation of a discharge activity based on the analysis of the measured electric field is widely studied in various geographical conditions. The technique of locating a CG flash is also widely researched. However, little is known on the discharge characteristics in equatorial and tropical regions, such as those associated with preliminary breakdown pulses (PBPs) and return strokes (RSs). Similarly, improvements can still be made on Lightning Locating Systems (LLSs), especially those using the time of arrival (TOA) technique. In particular, the operation of a three-station LLS to correctly locate a CG flash is very much desired to be proven. This work aims to obtain and analyse the lightning discharge characteristics in Malaysia. The work also proposes a new TOA based technique to correctly locate a CG flash using only three measuring stations. Measurements had been made in southern Malaysia using a purposely designed lightning detection system comprising a broadband antenna. A three-station TOA based LLS had also been implemented. The new TOA based technique was developed using the three-station LLS modelling in Matlab and artificial neural network (ANN). A discrete wavelet transform based technique was successfully developed to classify the discharge type. Self-Organizing Maps and Levenberg–Marquardt algorithms can identify the correct strike position with 2.5% error. The trained ANN engine was used to determine flash locations in a 400 km2 region. The three-station LLS gives superior results in terms of detection accuracy and efficiency when compared with those measured by Malaysian Meteorological Department. |
|---|